Talk:Hex key

From Wikipedia, the free encyclopedia
Jump to: navigation, search

History and nomenclature with refs[edit]

Added. — ¾-10 02:46, 10 November 2008 (UTC)


This article would benefit from an explanation of the origins "Allen", as these keys are also known. Any clues? Andrew Oakley (talk) 11:29, 30 March 2009 (UTC)

It already states that info in the History section. Wizard191 (talk) 13:27, 30 March 2009 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Internal spanners on the same principal were common in firearms back to the eighteenth century, for tightening the barrels on "turn-off" pistols, which instead of being loaded through the actual muzzle, were loaded by removing the barrel, loading pawder and ball, then screwing the barrel back on. They were usually octagonal rather than hexagonal, but anyway. A leading American manufacturer of pistols in the early nineteenth century was Ethan Allen. Could there be a connection? —Preceding unsigned comment added by (talk) 10:36, 3 February 2011 (UTC)

Interesting. I was not aware of these internal-wrenching octagonal sockets that predated the hex socket. My gut instinct suggests that the echo of the name (Ethan Allen vs Allen Manufacturing Company of Hartford, Connecticut) is just a coincidence. But it would be interesting to look into it. — ¾-10 23:29, 3 February 2011 (UTC)


Is "stoicastically" a real word? I suspect it's a misspelling of "stochastically", but even if it is, the sentence is opaque. Are the measurements truly random, or does it mean there some Monte Carlo method analysis? Someone who knows what this means should change it some the rest of us can read it.  Randall Bart   Talk  07:29, 21 April 2009 (UTC)

The proper word is stochastically, which encompasses the Monte Carlo method. I've corrected the article. Wizard191 (talk) 14:29, 21 April 2009 (UTC)
I think it was added by an anonymous contributor: see the diff here and, even with the correct spelling, hardly makes any sense. It looks more like a joke. If nobody comes up with a reason why it should be left there, I will erase that sentence. HumphUK (talk) 22:32, 22 January 2011 (UTC)
I've never understood it here, but I left it alone because I'm a statistical/mathematical layperson (i.e., when it comes to advanced applied math, I can only catch-as-catch-can and let the rest flow over my head), so I didn't have any basis for second-guessing it. But if anyone with higher math abilities can tell that it's misguided, I encourage removal. The burden of proof is on the original contributor to come back and re-add it with better explanation and/or refs if it's truly important to have here. Regards, — ¾-10 18:21, 23 January 2011 (UTC)
It doesn't make any sense to me either, so I removed it. Wizard191 (talk) 18:54, 23 January 2011 (UTC)

Tool standards[edit]

Does anyone have a reference to an industry standard for metric hex keys? The ASME standard only defines the 'inch' series; the metric data or source would be great. —Preceding unsigned comment added by (talk) 21:40, 11 February 2011 (UTC)

I searched at for "fasteners" and got 72 hits, some of which looked interesting. The most relevant title was ISO 272:1982 "Fasteners -- Hexagon products -- Widths across flats". Others that looked juicy were ISO 8992:2005 "Fasteners -- General requirements for bolts, screws, studs and nuts" and ISO 1891:2009 "Fasteners -- Terminology". But sadly the paywall threshold is CHF 44,00 high, so I can't afford to venture inside. God bless 'em, they need money to do what they do. Hope this helps. — ¾-10 23:01, 11 February 2011 (UTC)

8 mm drivers in 5/16 sockets[edit]

The article claims that due to the very close sizes 5/16in and 8mm keys are interchangable. Afaict imperial drivers are ok in metric sockets but metric drivers won't nessacerally fit in imperial sockets (having tried two on the imperial socket I have here one fits tightly and the another simply won't go in). I don't have a cite for this but the current statement in the article isn't cited either. Plugwash (talk) 13:44, 1 March 2011 (UTC)

Depends on the tolerances and allowances that the given wrench and fastener are made to. The closest pairs, with ≤0.05mm diff (i.e., 4mm–5/32, 19mm–3/4) pretty much always work. The slightly farther ones, with diffs from 0.05mm to 0.15mm (i.e., 8mm–5/16, 11mm–7/16, 16mm–5/8) often work. It seems that in some cases the makers use an undersize allowance from nominal to make sure that the "often" extends to "likely"—i.e., that it fits on the one end of the range at the expense of tolerable looseness on the other end. In fact on some lug wrenches this principle is carried to an almost comical level. On a four-ended brace, they mark each end nominally as 19mm–3/4, 21mm–13/16, 22mm–7/8, and 24mm–15/16. They're using the larger of each pair for the female socket and banking on the fact/theory (depending) that it'll work on a male of the lesser of each pair without heartbreak. It often does. The diff from 22mm to 7/8 is >0.20mm, and from 21mm to 13/16 is >0.35mm. Gettin pretty sloppy at that point. — ¾-10 23:58, 1 March 2011 (UTC)
Is this info even appropriate for an encyclopedia article on hex keys? I don't think so... Wizard191 (talk) 18:45, 9 March 2011 (UTC)
Meh, I would say keep it, but I wouldn't be upset if either of you guys chose to delete it. Ideally I think all such info brought together (interesting coincidental alignment points between SI and non-SI, and their practical uses) would make a nice section of an article somewhere in WP's coverage of applied metrology or of MRO (or maybe even just of wacky grade-school math games that cause kids to think about measures), and for all I know, such a home may already exist. In other words, I think it should be discoverable somewhere on WP, but I currently have no urge to insist on its inclusion in any particular article. — ¾-10 02:14, 10 March 2011 (UTC)
That's just it, I think it's outside the scope of this article. Maybe something can be added to inch or if you attack it from a different angle list of screw drives. Wizard191 (talk) 16:38, 16 March 2011 (UTC)


For the lay reader, shouldn't this say somewhere up front, in simple English, that hexagonal means these have six equal sides? Not everyone knows what a hexagon is (I hate to tell you). True, lower down it does say: "There are six contact surfaces between bolt and driver." I can figure out what this means, it is technically-correct. But it's jargony. What the heck is a 'contact surface'? I can think of all kinds of odd-looking shapes with six 'contact surfaces' -- dice cubes are the most common, with six contact surfaces, and there is in fact a dice-wrench [1] and even a company by that name.[2] Trestres (talk) 15:58, 1 July 2011 (UTC)

I think that we shouldn't talk down to people, and that we don't need to explain what a hexagon is. there's WP simple English for people who need it for what ever reason -- age, or not speaking English as a first language. Or maybe I'm missing the point, possibly you are just joking as the external links are to a video game site. --Keithonearth (talk) 00:03, 2 July 2011 (UTC)
I was typing a reply, too, given below. It mentions one of the same things that Keithonearth did too. Hope it's useful. — ¾-10 00:09, 2 July 2011 (UTC)
[Edit conflict]
Your goal (always working toward greater clarity) is an excellent one. My only counterargument is that we have to let the hyperlinks themselves do much of the work of supplying more information for those subsets of readers who need it. Why? Because otherwise you could add words to explain everything (for example, what a hexagon is) right here in this one article. It could get very awkward. It seems better to stick with a "drill-down" style of pedagogy, where anyone who needs to know more about hexagons can click through to the article on hexagons, while in the meantime, other readers who didn't need that level of hand-holding can just read the plaintext (anchor text), grasp it instantly, and move on, not having been slowed down at all by excess words. This is especially important because of how diverse the readership is; whatever you did to serve one subset of readers (for example, elementary school kids) could get in the way of another subset of readers (for example, college grads). One of the great things about the hypertext environment of the web, as opposed to the world of paper, is this selectiveness of "drill-down-ability" that helps anyone who needs it while also not cluttering things up for anyone else. Of course, one of the implications of the idea that Wikipedia is not paper is that if you were to print it out on paper, thus "flattening" it via the loss of information that results from losing the click-through-ability and drill-down-ability, lots of its pedagogical value would be lost or impeded (even though quite a bit would remain). But the answer to that is really just "True; but that's not a reason to avoid using the medium of hypertext to its fullest for pedagogical advantage. Most users will be using in the onscreen medium." This topic also has interesting implications in areas where pure science, applied science, and technology continue to revise humans' understanding of the world. If a neologism is moving gradually from science into layperson usage, what is the right ratio of (1) explaining the word in the plaintext (anchor text) versus (2) skipping that for brevity and just providing a link for anyone who needs to click through? Clearly neither one alone is always appropriate; but leaning too hard toward the first option, out of a tradition that evolved in the print-only era when there was no other choice, doesn't seem the right balance anymore. This shows up as a generational divide, too. (Fortunately, a mild, malleable one.) Kids born in 1996 are today [2011] 15-year-olds who have lived in the hypertext era their whole lives. They take to option 2 above very naturally, because their modes of reading have been shaped by it since their beginning. Meanwhile, someone who was already 40 years old by 1996 might have imbibed option 1 "as a given" for the first 40 years of their lives, an implicit default, and wonder why the world ever got away from it as much as it now has. The answer is merely that option 2 has a lot going for it, with the drill-down factor and the ability to serve multiple audiences simultaneously, with each user having a different click-through experience, but with all heading in the same direction pedagogically. Interesting to think about! One could argue, "Yeah, yeah, but even despite all that, the anchor text should still be more simplistic." This is often a valid enough point, but one way to address it is that there are actually two English-language Wikipedias: [regular] English Wikipedia, and Simple English Wikipedia. I would generally argue that we're getting the balance about right, overall; the "neediest" audiences (e.g., young children, EFL adults, native-speaker adults with learning disabilities) should be given the support they need at Simple English Wikipedia if they aren't able to use [regular] English Wikipedia effectively enough; although clarification of particular spots in the latter resource is, without doubt, always an ever-ongoing continuous improvement process. Regards, — ¾-10 00:09, 2 July 2011 (UTC)


A wrench (aka spanner) is a lever made to fit around a keyed shaft to provide external torque, while a key is a lever designed to fit within a keyed hole to provide internal torque. In other words, this is not a wrench. I have updated this in the first sentence of the article. Sp4i6 (talk) 15:56, 29 April 2014 (UTC)

Should "Hex key standard sizes" table include sizes available from manufacturers?[edit]

At present the table in the Hex key standard sizes is taken from the "Machinery's Handbook, 26th Edition, section "Fasteners", chapter "Cap and Set Screws", table 4 (p. 1601)." That seems like a decent WP:RS but I'm wondering if the table should in include other sizes and that we find another WP:RS similar to the Machinery's Handbook. Though I list them here I would not want to use the manufacturer pages as sources in the article.

Wikipedia 7/64 9/64 5/32 3/16 1/4 5/16 3/8 1/2 5/8 3/4 Source
Allen 0.028 0.035 0.05 1/16 5/64 3/32 7/64 1/8 9/64 5/32 3/16 7/32 1/4 5/16 3/8 7/16 1/2 9/16 5/8 3/4 7/8 1 1.25 1.5 1.75 2 [3]
Apex 1 0.05 1/16 5/64 3/32 7/64 1/8 9/64 5/32 3/16 7/32 1/4 5/16 3/8 [4]
Apex 2 3/32 7/64 1/8 9/64 5/32 3/16 7/32 1/4 9/32 5/16 3/8 1/2 9/16 5/8 3/4 [5]
Bondhus 0.028 0.035 0.05 1/16 5/64 3/32 7/64 1/8 9/64 5/32 3/16 7/32 1/4 5/16 3/8 7/16 1/2 9/16 5/8 3/4 7/8 1 1.25 1.5 1.75 2 [6]
Fastenal 0.028 0.035 0.05 1/16 5/64 3/32 1/8 5/32 3/16 7/32 1/4 5/16 3/8 1/2 9/16 3/4 [7]
Wiha 0.05 1/16 5/64 3/32 7/64 1/8 9/64 5/32 3/16 7/32 1/4 5/16 3/8 [8]

As you can see - some sizes not listed in the Wikipedia article are available from multiple sources implying they are common. Others, such as 9/32, seem to less common. The 0.028, 0.035, and 0.05" sizes do not translate to even metric sizes. --Marc Kupper|talk 05:15, 2 October 2014 (UTC)


I agree imbus is a misspelling but it's actually the correct and most-used form in Romanian. That's because there's a grammar rule that says m should come before p and b, instead of n. There are some exceptions to this, like 'Istanbul', but 'Istambul' is also sometimes accepted. Since inbus has been depersonalized (or debranded) the evolution has likely dictated that it follows the common grammar rules. It might be the same case in other languages as well. Disclaimer: I am not a linguist. 21:00, 20 September 2015 (UTC) (talk) 20:08, 20 September 2015 (UTC)Ligius